This thread has been entertaining, I must say, in ways both good and bad
. I was debating whether to participate or not; at certain points in its development, the place seemed overrun with hulking, beetle-browed mountain denizens out of a Grimm's fairy tale. Or maybe it's that when we get our dander up, and get red-faced and call each other the devil in self-righteously elevated falsetto, we all start to resemble one another, whether ubers or haters or something in between. We have met the enemy, and he is us.
But then I thought: ahhh, the heck with it. Might as well just dive in. As anyone who's ever been a little boy knows (which I realize is a tremendously un-PC generalization that excludes at least half the audience), sometimes there's nothing that clears the air better than a good old fashioned, mindless free-for-all.
When I was in grade school, we played a "game" that we called 'Russian Smuck', which consisted of a pack of boys chasing down and piling onto whoever had the ball, after which the ball was released so that someone else could have the honor of being the rabbit. There was nothing else to it
; I do not recall that there were any teams or that there was such a thing as a score. It was, nevertheless, tremendously popular for a while. I don't know where the name came from. My own theory is that this game, or some similarly atavistic variant, was the origin of rugby and American football.
With those meandering prefatory remarks out of the way, what are my own thoughts on the topic at hand? My general impression is that, as is often the case, the controversy stems from a lack of precision in defining the initial conditions of the question.
-What do we mean when we say "if she goes clean"? This question, I suggest, is not as simple as it seems. We all seem to vaguely agree that this hypothetical cleanliness ought to represent the best of which that particular skater is capable, element by element, component by component. But this begs the question: how is that skater's representative best, for the purpose of a patinageous soothsaying that has at least some veneer of rationality, to be defined and identified?
Is it enough to simply find the best example of an element, say, a triple flip, ever done by that skater in her career, and use the characteristics and scores for that singular example as the benchmark? And then do the same for all of the elements and components in a hypothetical program and sum them all up?
The issue, as most of us recognize, is that skater A may only have done a 3F of that quality once in her career, perhaps at a Nationals or a lower level comp, while skater B may have done it many times and on the highest stage. If probability and context are not taken into account and explicitly adjusted for, then the "answer" that results from the simple procedure above is that skater A and skater B are relatively equal "if clean". If that's the way one chooses to define the question, then the answer is, I freely grant, correct...And? What does that have to do with the price of tea in China?
Perhaps another example will make the point more clear. I personally have been known to hit golf drives more than 270 yards, stick approach shots to within a foot of the hole, and drain downhill, left-to-right 50 footers like they were on rails. On the face of it, it seems to me that I should have been one of the favorites for this year's US Open played on the somewhat short Merion course. The only caveat being that I can only do any of these things maybe one time in a thousand. And to do all of these things, and over the course of 72 holes, on a US Open set-up with millions watching, the probability is, I admit, somewhat lower than that. But those are just small details of consistency/execution, aren't they? If clean, I should be in a pairing with Justin Rose. Why aren't there any discussion board threads, I ask you, titled "What would be the leaderboard if Robeye, Tiger, Phil all go clean?"
The corollary that should logically follow is that it makes no sense, in any practical way, to say that, unadjusted for probability/consistency, Carolina or Mao are the equal of Yuna at this time. Weighting for consistency is not an optional feature; it is a fundamental condition, in my view, for the question (and answer) to make sense at all.
It is certainly not the way that things are viewed in other sports, or pragmatic pursuits such as businesses. If an eager staffer for a (profitable) bookmaker had decided: "Jayzus, that Alex Rodriguez can smack that ball six ways to Sunday. If he does that in that in the playoffs [which he has never been able to do], the Yankees are gonna win 6 World Series in a row; I'm gonna give 2-to-1 odds on that", that bookmaker would no longer be profitable, and the former staffer would probably be accepting bets in his office at the bottom of the East River, with his cement-shod feet propped up on his slippery desk.
One last example. If the Powerball amount ever gets to $1 billion, should I buy a ticket and try to persuade Bill Gates to accept it while giving me some like amount in cash, on the rationale that they are essentially equivalent, if I "go clean", as it were? While the differentials are obviously not comparable, I am clearly using it to highlight the logic, which is the same as for the initial framing of this question, or of arguments that Caro or Mao can currently be considered Yuna's equal at this time. I'm all for politically correct diplomacy and mutual good-feeling, but not at the cost of fact and reality.
-One logically coherent way to frame the question would be the following: start with any of the skaters, and construct a hypothetical program and target scores for each element and component. In parallel, a probability of success is derived for each element/component (at the associated target score) based on the historical data. From this, a probability of success for the entire program can be calculated.
For each of the other skaters, the question is: can a hypothetical program for them be constructed at the same probability of success, which equals or exceeds the first skater's target score (again, based on historical performance)?
Another variant would be: pick any probability of success that you want (for example, 50%, or 70% or 80%, etc.), then construct a program for each skater that maximizes the program score which, based on the historical data, meets that probability of success.
-Long story short: I am of the view that at any desired level of probability one chooses, there is a possible Yuna program that outscores that of Caro or Mao.
What is the probability that Caro or Mao will go clean with some of the hypothetical programs discussed in this thread? I haven't done the calculations, but is it even 5%?
The conceptual objection is not the 5% (although that's a pretty low level that calls into question the practical relevance of the exercise), it is that we are utilizing Caro and Mao programs with very low probabilities of success, while arbitrarily, it seems to me, limiting Yuna to a program with, relatively speaking, an extremely high probability of success.
Why should that necessarily be so? In any competitive endeavor, whether in evolutionary biology or business or geopolitics or athletics, one's plans are not created in a vacuum, but in response to situational exigencies. The program arms race among the ladies is actually a very good example of this. The reason that Mao has evolved such a high BV program is in response to Yuna's huge advantage in GOE; Caro's program has also evolved in the past year (and may continue to evolve in the next) because of a similar impetus, that is, the need to counter Yuna's advantages.
That they are pursuing what are, historically speaking, relatively dicey program propositions for them, in order to keep alive any hope of Olympic gold, is actually the measure of the gulf between Caro and Mao on the one hand, and Yuna on the other, at the current time. In fact, it was Caro who commented after Worlds that Yuna "is on another planet". One underestimates the intelligence of skaters if we assume that they don't have a very clear-eyed picture of what's what.
This also goes the other way, however. Yuna chooses to do a 6-triple program for the simple and persuasive reason that, on a probability-adjusted (or equivalent-probability) basis, she can be very comfortable that she does not need to do more to win, and win going away, if she executes well. The error on the part of viewers, IMO, is in thinking that this represents the limit of her possible programs.
If Yuna felt that she needed to, are there any hypothetical programs with a probability of success (as calculated using some reasonable method based on historical data) that matches the relatively low probability thresholds for the hypothetical Caro/Mao programs (5%, as per my generalized example above; I reiterate that I have not calculated the actual probabilities), but which materially exceed their hypothetical point totals? In my view, the answer to this is "yes". If anything, my own view is that the prospects of success for Yuna to add a 3lo to her programs, or two triple-triples in the LP, etc.(and are therefore accretive to her points total), are at least as good, if not better, than the prospects that Caro or Mao will go clean in their respective low-probability hypothetical programs.
I feel obligated to add the following standard disclaimer to avoid misunderstanding and flaming: Probability is not certainty, particularly in sports, and particularly in the somewhat one-off type of format that is skating. The "on any given Sunday" element quite clearly exists, and thus, while a descending list of favorites can be created using techniques of quantitative analysis, all of the skaters, particularly the top trio, still have everything to play for.
(Yes, I apologize for the length of post; members who have been around for a while will know that as hard as I try to keep myself in check, sometimes one just busts out, and will therefore forgive.
